Download Unsaturated Fatty Acids Increase Plasminogen Activator Inhibitor

Unsaturated Fatty Acids Increase Plasminogen Activator
Inhibitor-1 Expression in Endothelial Cells
Lennart Nilsson, Cristina Banfi, Ulf Diczfalusy, Elena Tremoli, Anders Hamsten, Per Eriksson
Downloaded from http://atvb.ahajournals.org/ by guest on June 18, 2017
Abstract—In vivo studies have demonstrated a strong positive correlation between plasma very low density lipoprotein
(VLDL) triglyceride and plasma plasminogen activator inhibitor-1 (PAI-1) activity levels. Furthermore, VLDL has been
shown to induce PAI-1 secretion from cultured endothelial cells. In contrast, no or variable effects on PAI-1 secretion
have been reported for native low density lipoprotein. It could be speculated that fatty acids derived from VLDL
triglycerides are the actual mediators, resulting in an enhanced secretion of PAI-1. In the present study, we have
analyzed the effects of both saturated and unsaturated fatty acids on PAI-1 expression and secretion by endothelial cells.
Addition of 0 to 50 mmol/L of either palmitic acid or stearic acid had no effect on PAI-1 secretion from human umbilical
vein endothelial cells or EA.hy926 cells. In contrast, addition of oleic acid, linoleic acid, linolenic acid, and
eicosapentaenoic acid resulted in a significant increase in PAI-1 secretion from both cell types. Northern blot analysis
of PAI-1 mRNA levels was in agreement with these findings. Transfection experiments demonstrated that addition of
linolenic acid and eicosapentaenoic acid significantly increased PAI-1 transcription. The fatty acid response region was
localized to a previously described VLDL-inducible region of the PAI-1 promoter. Electromobility shift assays
demonstrated that unsaturated fatty acids induced the same complex as did VLDL, whereas saturated fatty acids had no
effect. Furthermore, it was demonstrated that the activation procedure did not involve fatty acid oxidation to any
significant extent. In conclusion, the present study demonstrates that unsaturated fatty acids increase PAI-1 transcription
and secretion by endothelial cells in vitro. The effect appears to be mediated by a previously described VLDL-inducible
transcription factor. (Arterioscler Thromb Vasc Biol. 1998;18:1679-1685.)
Key Words: PAI-1 n fatty acids n promoter n endothelial cells n VLDL
P
triglyceride content. Furthermore, it could be speculated that
fatty acids derived from VLDL triglycerides are the actual
mediator, resulting in an enhanced release of PAI-1. Indeed,
in vitro experiments have demonstrated that docosahexaenoic
acid and dihomogamma linolenic acid induce PAI-1 mRNA
in HUVECs14 and that linoleic acid enhances PAI-1 secretion
from HepG2 cells.10 In agreement with the in vitro data,
administration of n-3 fatty acids in vivo has resulted in
increased plasma PAI-1 activity.15–18
Recently, a VLDL response element was identified in the
promoter region of the PAI-1 gene locus that mediates
VLDL-induced PAI-1 transcription in endothelial cells.19 A
VLDL-inducible transcription factor binds directly downstream of the common 4G/5G polymorphic site in the PAI-1
promoter. Competitive binding between the VLDL-inducible
transcription factor and the 5G allele–specific transcriptional
repressor may explain the allele-specific differences in the
association between plasma triglycerides and PAI-1 activity
observed in non–insulin-dependent diabetic patients and in
patients with coronary artery disease.20 –22
lasminogen activator inhibitor-1 (PAI-1), the fast-acting
inhibitor of plasminogen activators, is the principal regulator
of the endogenous fibrinolytic enzyme system. Low fibrinolytic
capacity has been associated with manifest coronary heart
disease and increased risk of recurrent major cardiovascular
events in patients with a history of cardiovascular disorders.1
Both environmental and genetic factors contribute to determine plasma PAI-1 activity. Among PAI-1 associations
with established risk indicators for coronary heart disease, the
relation with VLDL has been analyzed extensively. In vivo
studies consistently have demonstrated a strong positive
correlation between the plasma VLDL triglyceride and PAI-1
activity levels.2–5 In vitro, VLDL has been shown to induce a
concentration-dependent increase in the PAI-1 secretion from
cultured human umbilical vein endothelial cells (HUVECs)6 – 8
and HepG2 cells.7,9 Addition of a triglyceride-rich emulsion
also resulted in an enhanced secretion of PAI-1 by HepG2
cells.10 In contrast, no or variable effects on PAI-1 secretion
by cultured cells have been reported for native LDL.6,8,11–13
Thus, the effects of lipoproteins could be influenced by their
Received January 20, 1998; revision accepted April 17, 1998.
From the Atherosclerosis Research Unit, King Gustaf V Research Institute, Department of Medicine, Karolinska Hospital (L.N., A.H., P.E.), and the
Department of Medical Laboratory Sciences and Technology, Division of Clinical Chemistry, Huddinge University Hospital, Karolinska Institute (U.D.),
Stockholm, Sweden; and the Institute of Pharmacological Sciences, University of Milan (C.B., E.T.), Italy.
Correspondence to Per Eriksson, King Gustaf V Research Institute, Karolinska Hospital, S-171 76 Stockholm, Sweden. E-mail
[email protected]
© 1998 American Heart Association, Inc.
Arterioscler Thromb Vasc Biol. is available at http://www.atvbaha.org
1679
1680
Fatty Acid Induction of PAI-1
In the present study, we have analyzed the effects of both
saturated and unsaturated fatty acids on PAI-1 expression and
secretion by endothelial cells. Furthermore, the molecular
mechanism whereby fatty acids stimulate PAI-1 secretion has
been studied and linked to the VLDL activation pathway.
scavenger, was used in some experiments to prevent fatty acid
oxidation in the medium. The EA.hy926 cells were incubated with
20 mmol/L of Trolox for 30 minutes before addition of fatty
acid–BSA complexes and subsequent incubation for 14 hours before
collecting the medium.
Northern Blot Analysis
Methods
Cell Culture
Downloaded from http://atvb.ahajournals.org/ by guest on June 18, 2017
HUVECs were isolated from umbilical cords obtained at normal
deliveries. The umbilical vein was cannulated and perfused with 50
mL PBS to remove any blood, whereafter the vein was filled with 20
mL 0.1% collagenase dissolved in PBS and incubated for 15 minutes
at 37°C. The collagenase solution was drained from the cord and
collected, and the cord was flushed gently with 20 mL PBS, which
was added to the collagenase solution. The cells in these pooled
solutions were recovered by centrifugation at 200g for 5 minutes and
seeded out on 9-cm culture dishes in M199 medium with 20% FCS,
antibiotic/antimycotic (Sigma Chemical Co), and 25 mg/mL endothelial cell growth supplement (Sigma Chemical Co). The cells were
subcultured onto 0.2% gelatin (in PBS)– coated dishes when confluent. Cells from pooled multiple cords were used for experiments
until the fourth passage. The endothelium-derived cell line EA.hy926
(a kind gift from Dr C.-J.S. Edgell, University of North Carolina,
Chapel Hill, NC) was cultured in DMEM with high glucose
supplemented with 10% FCS, HAT (100 mmol/L hypoxanthine,
0.4 mmol/L aminopterin, and 16 mmol/L thymidine), penicillin, and
streptomycin as described.23
VLDL Preparation
VLDL for incubation with HUVECs was prepared by density
gradient ultracentrifugation.24 The endotoxin content in the VLDL
preparations was tested using a Limulus amebocyte lysate assay
(COATEST Endotoxin, Endosafe Inc). Endotoxin levels were shown
to be ,0.1 ng/mg protein.
Preparation of Fatty Acid–BSA Complexes
Fatty acid–BSA complexes were prepared essentially according to
the method of Spector and Hoak.25 In brief, 25 mg of fatty acids
(16:0, 18:0, 18:1, 18:2, 18:3, and 20:5; Sigma Chemical Co) was
dissolved in 7.5 mL hexane, and 800 mg Celite was added. The
solvent was removed under N2 by continuous magnetic stirring.
When the solvent had evaporated completely, fatty acid–free BSA
(25 mL of 0.25 mmol/L; Sigma Chemical Co) was added. The
mixture was stirred for 1 hour at room temperature with N2
constantly passing over the surface. After centrifugation at 800g for
5 minutes, the supernatants were decanted carefully. Samples containing fatty acid–BSA complexes were filtered and stored in
aliquots under N2 at 220°C.
13-Hydroperoxy-9,11-octadecadienoic acid (13-OOH-18:2) was
synthesized as described.26 In brief, linoleic acid was incubated with
soybean lipoxygenase at 0°C in borate buffer at pH 9.0. The product
was purified by silicic acid column chromatography, and the purity
was determined by high-performance liquid chromatography.
Semiconfluent cultures of EA.hy926 cells were preincubated for 8 to
10 hours in DMEM containing 1% charcoal-treated FCS before
incubation with the fatty acids. Total RNA from the EA.hy926 cells
was isolated according to the Rneasy handbook (Qiagen). Northern
blotting and hybridization on DuPont GeneScreen Plus nylon membranes (NEN Research Products) were performed according to the
manufacturer’s protocol. Blots were hybridized with 106 cpm/mL
[32P]dCTP-labeled SfiI and BglII fragment (1255 bp) of the cDNA
for PAI-1 (courtesy of Dr T. Ny, Department of Medical Biochemistry and Biophysics, University of Umeå, Umeå, Sweden).
Transfection Assay
EA.hy926 cells were transfected using a calcium phosphate precipitation method as described by Sambrook et al.28 pRSV-galactosidase
control vector (Promega) was cotransfected as an internal control.
The construction of the PAI-1 CAT plasmids has been described
elsewhere.19 The 4G-PAI-pCAT construct comprises the human
PAI-1 sequences 2804 to 17. The truncated promoter constructs,
2708-PAI-pCAT and 2609-PAI-pCAT, were constructed from the
4G-PAI-pCAT as described.19 The 4G-9DEL-PAI-pCAT plasmid
was constructed using the Altered sites II in vitro mutagenesis
system (Promega). A 9-bp deletion was introduced just downstream
of the 4G/5G polymorphic site of the 4G-PAI-pCAT construct.19 The
cells were transfected at 80% to 90% confluence. One to 3 hours
before transfection, the dishes received fresh complete medium.
Cells were incubated for 4 hours with calcium phosphate–precipitated DNA (15 mg plasmid per 90-mm dish). After a 2-minute 15%
(vol/vol) glycerol shock, fresh medium containing 1% charcoaltreated FCS and fatty acids was added, and the cells were harvested
for transient expression 16 to 18 hours later. CAT activity was
analyzed subsequently according to Sambrook et al.28
Electromobility Shift Assay (EMSA)
Nuclear extracts were prepared according to Alksnis et al.29 All
buffers were supplemented freshly with 0.7 mg/mL leupeptin, 16.7
m g/mL aprotinin, 0.5 mmol/L PMSF, and 0.33 m L/mL
2-mercaptoethanol. The protein concentration in the extracts was
estimated by the method of Kalb and Bernlohr.30 For EMSA, a
double-stranded oligonucleotide comprising the 2675 to 2653
region of the PAI-1 promoter was designed. Semiconfluent cultures
of HUVECs were incubated for 8 to 10 hours in M199 medium
containing 1% charcoal-treated FCS. This was followed by an 8-hour
incubation with fatty acids before the preparation of the cell extracts.
Incubation conditions for EMSA were as described.19 To test for
specific interaction of the VLDL- and fatty acid–induced factor,
nonlabeled specific and nonspecific probes were used as competitors19 (data not shown).
Statistical Methods
Determination of PAI-1 Protein Secretion
Semiconfluent cultures of HUVECs or EA.hy926 cells were incubated for 8 to 10 hours in M199 or DMEM medium, respectively,
containing 1% charcoal-treated FCS. This incubation was followed
by a 14-hour incubation with fatty acids added in the same type of
medium. After collecting the conditioned medium and centrifugation
at 9000g for 5 minutes, the PAI-1 protein concentration in the
medium was quantified using an ELISA (TintELIZE PAI-1,
Biopool) that detects active and inactive (latent) forms of PAI-1, as
well as tissue plasminogen activator/PAI-1 complexes. The cells
were either trypsinized and counted or lysed with 0.01 M NaOH
followed by measurement of total protein.27 PAI-1 secretion was
expressed as percentage of control (vehicle containing the same
amount of BSA solution added). Trolox (Fluka), a peroxyl radical
Differences in continuous variables between 2 groups were tested by
an unpaired Student t test. Data are mean6SD.
Results
Effects of Fatty Acids on PAI-1 Secretion and
mRNA Levels in Endothelial Cells
Fatty acids were incubated with HUVECs or with the
HUVEC-derived cell line EA.hy926, and the PAI-1 secreted
into the medium was measured using ELISA. Palmitic (16:0),
stearic (18:0), oleic (18:1), linoleic (18:2), linolenic (18:3),
and eicosapentaenoic (EPA) (20:5) acids were complexed
with BSA and incubated for 14 hours with the cells before
Nilsson et al
Downloaded from http://atvb.ahajournals.org/ by guest on June 18, 2017
Figure 1. Fatty acid induction of PAI-1 secretion from HUVECs
(A) and EA.hy926 cells (B). Fatty acids were incubated with the
cells for 14 hours, whereafter PAI-1 contents of culture medium
were determined by ELISA. Results (mean6SD) are given as
percentage of control. Results were derived from 4 to 8 experiments, all performed in triplicate.
collecting the conditioned medium. As shown in Figure 1A
and 1B, the effects of the fatty acids on PAI-1 secretion from
HUVECs and EA.hy926 cells were similar. Palmitic acid or
stearic acid (0 to 50 mmol/L) had no major effect on PAI-1
secretion from either HUVECs or EA.hy926 cells. A small
increase in PAI-1 release from HUVECs was obtained with
10 to 25 mmol/L stearic acid (Figure 1A). In contrast, oleic
acid, linoleic acid, linolenic acid, and EPA showed a dosedependent effect on PAI-1 secretion from both cell types.
Addition of 50 mmol/L of either oleic acid, linoleic acid,
linolenic acid, or EPA resulted in a 40% (P,0.001), 59%
(P,0.001), 60% (P,0.001), and 54% (P,0.001) increase in
PAI-1 secretion from HUVECs, respectively, and in a 35%
(P,0.001), 42% (P,0.001), 55% (P,0.001), and 62%
(P,0.001) increase in PAI-1 secretion from EA.hy926 cells,
respectively (Figure 1A and 1B). The basal secretion of
PAI-1 from HUVECs and EA.hy926 cells was 100 to 120
ng/105 cells and 20 to 30 ng/105 cells, respectively.
To test whether oxidation of the unsaturated fatty acids was
implicated in their effect on PAI-1 secretion, the peroxyl
radical scavenger Trolox (20 mmol/L) was incubated with
EA.hy926 cells before addition of 50 mmol/L linolenic acid.
No effect of the antioxidant was demonstrated on the linolenic acid–mediated induction of PAI-1 secretion (data not
shown). As a positive control for the activity of Trolox, it was
demonstrated that Trolox decreased the UV-induced mobility
change of LDL on agarose gel electrophoresis. We also
studied the effect of 13-OOH-18:2 on PAI-1 secretion from
EA.hy926 cells. Assuming that a maximum of 10% autooxidation of the fatty acid, 0 to 5 mmol/L of 13-OOH-18:2,
November 1998
1681
Figure 2. A, Northern blot analysis of PAI-1 mRNA recovered
from EA.hy926 cells after an 8-hour incubation with 50 mmol/L
of linolenic (18:3), palmitic (16:0), or oleic (18:1) acid. Total RNA
(5 mg) was hybridized with labeled cDNA probe for PAI-1. C
indicates vehicle containing same amount of BSA solution
added as control. The corresponding blotting filters stained with
methylene blue showing the 28S and 18S ribosomal RNAs demonstrate that approximately equal amounts of RNA were loaded.
B, Quantification of 3 experiments. Results (mean6SD) are
given as percentage of control. *P,0.05; **P,0.01.
was incubated with EA.hy926 for 14 hours. No effect on
PAI-1 secretion was detected with any of the 13-OOH-18:2
concentrations used. Addition of 0.1, 0.5, 1.0, or 5.0 mmol/L
of the peroxidized linoleic acid resulted in 96612%, 9766%,
9969%, and 99612% of the control PAI-1 antigen secretion,
respectively (mean6SD of 3 experiments performed in
triplicate).
Because fatty acids increased the secretion of PAI-1 from
HUVECs and EA.hy926 cells in a similar fashion, RNA and
transfection analyses (shown below), experiments that require
many cells, were performed only in EA.hy926 cells. Northern
blot analysis of mRNA levels was in agreement with the
finding that unsaturated fatty acids increase the secretion of
PAI-1 by EA.hy926 cells. Figure 2 shows representative
Northern blot analyses of the mRNA recovered from
EA.hy926 cells after stimulation with 50 mmol/L of either
palmitic, oleic, or linolenic acid. Linolenic and oleic acid had
a significant effect on PAI-1 mRNA levels. Both the 3.2-kb
(P,0.01) and the 2.2-kb (P,0.05) PAI-1 transcripts were
increased. In contrast, 50 mmol/L of palmitic acid did not
have an effect on PAI-1 mRNA levels (Figure 2). The
stimulatory effect on PAI-1 mRNA by linolenic acid was
detected after 2 hours (Figure 3).
Fatty Acid Activation of PAI-1 Transcription
A transfection assay was performed using an 804-bp fragment of the PAI-1 promoter coupled to a CAT gene. As
1682
Fatty Acid Induction of PAI-1
Downloaded from http://atvb.ahajournals.org/ by guest on June 18, 2017
Figure 3. A, PAI-1 mRNA recovered from EA.hy926 cells after
2- to 8-hour incubation with 0 to 50 mmol/L linolenic (18:3) acid.
Total RNA (5 mg) was hybridized with a labeled cDNA probe for
PAI-1. Vehicle containing the same amount of BSA solution was
added as control. Corresponding blotting filters stained with
methylene blue showing the 28S and 18S ribosomal RNAs demonstrate that approximately equal amounts of RNA were loaded.
B, Quantification of the above autoradiogram. PAI-1 mRNA is
given as percentage of control.
demonstrated in Figure 4, addition of palmitic (Figure 4A) or
stearic (Figure 4B) acid did not have any effect on PAI-1
transcription. In contrast, both linolenic acid (Figure 4C) and
EPA (Figure 4D) significantly increased PAI-1 transcription
(P,0.01 and P,0.01, respectively). To localize the fatty
acid–responsive region(s) in the PAI-1 promoter, we used
several truncations of the promoter. As demonstrated in
Figure 5, both the 2804-PAI-pCAT (Figure 5A) and the
2708-PAI-pCAT (Figure 5B) promoter constructs responded
significantly to addition of 50 mmol/L EPA, whereas the
2609-PAI-pCAT (Figure 5C) promoter construct did not.
This implies that the response element is located between
positions 2609 and 2708 of the PAI-1 promoter. This region
contains the previously identified VLDL response element
located between residues 2672 and 2657. To determine
whether the same response element in the PAI-1 promoter is
involved in both VLDL- and fatty acid–mediated induction of
PAI-1 transcription, we performed a transfection assay using
a promoter construct with a 9-bp deletion (residues 2670 to
2662) of the VLDL response element (Figure 6). This
deletion previously has been shown to eliminate the VLDL
responsiveness of the PAI-1 promoter. As shown in Figure
6B and 6C, use of this promoter construct completely
abolished the EPA-mediated induction of PAI-1 transcription.
Unsaturated Fatty Acids Increase the Binding of a
VLDL-Inducible Transcription Factor to the
PAI-1 Promoter
Because the transfection assays indicated that the recently
characterized VLDL-inducible transcription factor could be
involved in the fatty acid–mediated activation of PAI-1
transcription, EMSAs were performed. Nuclear extracts derived from HUVECs treated with fatty acids for 8 hours were
incubated with a probe containing the 2675 to 2653 region
of the PAI-1 promoter. As shown in Figure 7, the unsaturated
Figure 4. Unsaturated fatty acids activate transcription from the
PAI-1 promoter in EA.hy926 cells. A promoter construct containing 804 residues of the PAI-1 promoter coupled to a CAT
gene was transfected transiently into EA.hy926 cells. Palmitic
acid (16:0) (A), stearic acid (18:0) (B), linolenic acid (18:3) (C), or
EPA (20:5) (D) (50 mmol/L) was incubated with the cells for 16 to
18 hours. Bars indicate mean6SD, and PAI-1 transcription rate
is given as percentage of control after correction for
b-galactosidase activity. Results are based on 3 experiments
performed in duplicate. **P,0.01.
fatty acids induced the same complex as did 75 mg/mL of
VLDL. Neither stearic acid (Figure 7) nor palmitic acid
(Figure 8) had any effect on the binding of the VLDLinducible factor. In contrast, oleic acid, linoleic acid, and
EPA (Figures 7 and 8) increased the binding of the VLDLinducible factor.
Discussion
In the present study, we demonstrated that unsaturated fatty
acids increase PAI-1 transcription and secretion by endothelial cells in vitro. The effect appears to be mediated by a
previously described VLDL-inducible transcription factor. To
the best of our knowledge, this study is the first to demonstrate a mechanism by which fatty acids can modulate PAI-1
transcription positively.
Both VLDL and unsaturated fatty acids induced the binding of the same transcription factor to the PAI-1 promoter in
Nilsson et al
November 1998
1683
Downloaded from http://atvb.ahajournals.org/ by guest on June 18, 2017
Figure 5. Fatty acid response element is located within the
2609 to 2708 region of the PAI-1 promoter. Promoter constructs containing 804 (A), 708 (B), or 609 (C) residues of the
PAI-1 promoter coupled to a CAT gene were transfected transiently into EA.hy926 cells and induced by 50 mmol/L EPA
(20:5). Bars indicate mean6SD, and PAI-1 transcription rate is
given as percentage of control after correction for
b-galactosidase activity. Results are based on 3 experiments
performed in duplicate. *P,0.05; ***P,0.001.
Figure 6. Fatty acid response element coincides with VLDL
response element. Promoter constructs containing 804 (A) or
804 residues with a 9-bp deletion of VLDL response element (B)
of PAI-1 promoter coupled to CAT gene were transfected transiently into EA.hy926 cells and induced by 50 mmol/L EPA
(20:5). Bars indicate mean6SD, and the PAI-1 transcription rate
is given as percentage of control after correction for
b-galactosidase activity. Results are based on 3 experiments
performed in duplicate. **P,0.01. C, Example of autoradiography of thin-layer chromatography analysis of CAT assay using
1-deoxychloramphenicol (Amersham) as substrate.
vitro. Fatty acids derived from VLDL triglycerides also may
function as activators of the factor in vivo. The fatty acid
composition of the VLDL used in this study, unfortunately, is
not available. In a previous study, the weight percentages of
16:0, 18:0, 18:1, 18:2, and 20:5 in VLDL from fasting
subjects were 32.5%, 3.8%, 38.6%, 16.9%, and 0.2%, respectively (E.T. et al, unpublished data, 1995). The concentrations
of fatty acids used in the present study are in accordance with
the concentration of nonesterified fatty acids found in serum.
As demonstrated by Crofts et al,31 the concentrations of
nonesterified 16:0, 18:0, 18:1, and 18:2 were 74, 47, 68, and
36 mmol/L, respectively, in serum of fasting control subjects.
Several fatty acid–inducible transcription factors have been
described. Among these, the peroxisomal proliferator activator receptor (PPAR) family has been studied extensively.32
Members of the PPAR family are ligand-dependent transcription factors that bind to their cognate ligand with high affinity
and then activate gene transcription through binding to a
specific hormone response element in the promoter region of
the target gene (a peroxisome proliferator activator response
element [PPRE]). The VLDL/fatty acid response element in
the PAI-1 promoter shows some homology with a PPRE.19
However, the sequence homology between the VLDL/fatty
acid response element and a PPRE is only moderate, with a
67% homology with each hexamer of the site. A variety of
fatty acids, both saturated and unsaturated, activate PPAR in
vitro,33 and it has been proposed that fatty acids are the natural
ligands of PPARs.34,35 Furthermore, unsaturated fatty acids
recently have been demonstrated to bind PPAR in vitro.36 The
fact that saturated fatty acids do not activate the VLDLinducible transcription factor suggests that this factor is not
identical with any of the 3 subtypes of PPAR known to date.
However, PPARs belong to a rapidly growing family of
“orphan” receptors, and it is likely that new members will
appear. We are now in the process of cloning the VLDL/fatty
acid–inducible transcription factor.
Because lipoproteins are readily oxidized when incubated
with cultured cells in vitro, it can be envisaged that oxidized
fatty acids are the mediators of the VLDL/fatty acid– enhancing effect on PAI-1. However, as no inhibitory effect on
PAI-1 secretion was obtained with Trolox and 13-OOH-18:2
did not induce PAI-1 secretion, it seems reasonable to assume
that the activation procedure did not involve fatty acid
oxidation to any significant extent. We cannot exclude the
1684
Fatty Acid Induction of PAI-1
Figure 7. Fatty acid induction of the VLDL-inducible transcription factor. Representative autoradiogram of 2 EMSA experiments using protein extracts derived from HUVECs that had
been incubated with BSA-vehicle (lane 2), 25 to 50 mmol/L of
either stearic (18:0) (lanes 3 and 4), linoleic (18:2) (lanes 5 and
6), or linolenic (18:3) acid (lanes 7 and 8), VLDL-vehicle (lane 9),
and 75 mg/mL VLDL (lane 10), and bound to the 2675/2653
PAI-1 probe. Lane 1 shows probe in the absence of nuclear
extract. F indicates free probe; arrow, VLDL-inducible factor.
Downloaded from http://atvb.ahajournals.org/ by guest on June 18, 2017
possibility that intracellular oxidation of fatty acids is mediating the stimulatory effect and that the negative effect of
13-OOH-18:2 compared with linoleic acid is a result of an
altered uptake by the cells. The finding that oleic acid, a fatty
acid that shows very limited proneness to oxidation, enhanced
PAI-1 secretion to a similar extent as polyunsaturated fatty
acids, further supports the interpretation that the effect of
unsaturated fatty acids on PAI-1 secretion is not secondary to
oxidation. The finding that the fatty acid–mediated increase
of PAI-1 mRNA levels already occurs after 2 hours also
supports this notion.
An abundance of studies have confirmed the positive
association between plasma triglycerides and plasma PAI-1
activity.2 Reduction of hypertriglyceridemia also has been
indicated to improve the fibrinolytic potential.37–39 A concomitant reduction of body weight, serum triglycerides, and
plasma PAI-1 activity has been reported in several studies.40 – 42 Fish oils or long-chain, polyunsaturated n-3 fatty
acids have been shown to lower triglyceride concentrations in
hypertriglyceridemia when given in high concentrations.
However, the fatty acid intervention studies suggest that, in
addition to the lowering of the triglyceride levels, there could
be a positive and direct effect of the n-3 fatty acids on PAI-1
expression. For example, supplementation of the diet with n-3
fatty acids reduced the triglyceride level but increased plasma
PAI-1 activity in non–insulin-dependent diabetes mellitus
patients15 or patients undergoing coronary bypass surgery.17
Intake of n-3 polyunsaturated fatty acids or fish oils has also
been associated with increased plasma PAI-1 activity in
healthy individuals.16,18 Taken together, these clinical data
support the notion that n-3 polyunsaturated fatty acids have a
direct and positive effect on PAI-1 secretion also in vivo. The
present study, along with 2 previous reports, demonstrate that
this is, indeed, the case in vitro. Docosahexaenoic acid
increased PAI-1 mRNA levels in HUVECs,14 and linoleic
acid increased PAI-1 secretion from HepG2 cells.10 Here, we
show that unsaturated fatty acids, including n-3 fatty acids,
increase PAI-1 secretion, mRNA levels, and PAI-1 transcription in endothelial cells. However, it should be noted in this
context that there are also some clinical studies showing an
association between n-3 fatty acid intake and decreased
plasma PAI-1 activity. Lopez-Segura et al43 showed that
consumption of a diet rich in monounsaturated fatty acids
resulted in a significant decrease in both plasma PAI-1
activity and antigen in healthy individuals. Furthermore, the
triglyceride levels were not affected by the dietary treatment.
Similarly, n-3 polyunsaturated fatty acids recently have been
shown not to affect plasma PAI-1 activity in patients with
hypertension.44 These reservations notwithstanding, the in
vitro findings presented here suggest that unsaturated fatty
acids have a direct enhancing effect on PAI-1 synthesis and
that this could explain the apparent discrepancy between
increased plasma PAI-1 activity and decreasing triglyceride
levels during n-3 fatty acid supplementation in vivo.
Acknowledgments
This project was supported by grants from the Swedish Medical
Research Council (8691 and 11807), the Swedish Heart-Lung
Foundation, the European Commission (HIFMECH study, contract
BMH4-CT96-0272), the Marianne and Marcus Wallenberg Foundation, the King Gustaf V and Queen Victoria Foundation, the King
Gustaf V 80th Birthday Foundation, and the Professor Nanna Svartz
Foundation. We are grateful to Barbro Burt for excellent technical
assistance.
References
Figure 8. Fatty acid induction of a VLDL-inducible transcription
factor. Representative autoradiogram of 2 EMSA experiments
using protein extracts derived from HUVECs that had been
incubated with BSA vehicle (lane 2), 25 to 50 mmol/L of oleic
acid (18:1) (lanes 3 to 4), palmitic acid (16:0) (lanes 5 to 6), or
EPA (20:5) (lanes 7 to 8), and bound to the 2675/2653 PAI-1
probe. Lane 1 shows probe in the absence of nuclear extract. F
indicates free probe; arrow, VLDL-inducible factor.
1. Hamsten A, Eriksson P. Fibrinolysis and atherosclerosis. Baillieres Clin
Haematol. 1995;8:345–363.
2. Hamsten A, Wiman B, de Faire U, Blombäck M. Increased plasma levels
of a rapid inhibitor of tissue plasminogen activator in young survivors of
myocardial infarction. N Engl J Med. 1985;313:1557–1563.
3. Juhan-Vague I, Vague P, Alessi MC, Badier C, Valadier J, Aillaud MF,
Atlan C. Relationships between plasma insulin triglyceride, body mass
index, and plasminogen activator inhibitor 1. Diabetes Metab. 1987;13:
331–336.
4. Mehta J, Mehta P, Lawson D, Saldeen T. Plasma tissue plasminogen
activator inhibitor levels in coronary artery disease: correlation with age
and serum triglyceride concentrations. J Am Coll Cardiol. 1987;9:
263–268.
5. Asplund-Carlson A, Hamsten A, Wiman B, Carlson LA. Relationship
between plasma plasminogen activator inhibitor-1 activity and VLDL
triglyceride concentration, insulin levels and insulin sensitivity: studies in
randomly selected normo- and hypertriglyceridaemic men. Diabetologia.
1993;36:817– 825.
Nilsson et al
Downloaded from http://atvb.ahajournals.org/ by guest on June 18, 2017
6. Stiko-Rahm A, Wiman B, Hamsten A, Nilsson J. Secretion of plasminogen activator inhibitor-1 from cultured human umbilical vein endothelial
cells is induced by very low density lipoprotein. Arteriosclerosis 1990;
10:1067–1073.
7. Mussoni L, Mannucci L, Sirtori M, Camera M, Maderna P, Sironi L,
Tremoli E. Hypertriglyceridemia and regulation of fibrinolytic activity.
Arterioscler Thromb. 1992;12:19 –27.
8. Kaneko T, Wada H, Wakita Y, Minamikawa K, Nakase T, Mori Y,
Deguchi K, Shirakawa S. Enhanced tissue factor activity and plasminogen activator inhibitor-1 antigen in human umbilical vein endothelial cells
incubated with lipoproteins. Blood Coagul Fibrinolysis. 1994;5:385–392.
9. Sironi L, Mussoni L, Prati L, Baldassarre D, Camera M, Banfi C, Tremoli
E. Plasminogen activator inhibitor type-1 synthesis and mRNA
expression in HepG2 cells are regulated by VLDL. Arterioscler Thromb
Vasc Biol. 1996;16:89 –96.
10. Banfi C, Risé P, Mussoni L, Galli C, Tremoli E. Linoleic acid enhances
the secretion of plasminogen activator inhibitor type 1 by HepG2 cells. J
Lipid Res. 1997;38:860 – 869.
11. Latron Y, Chautan M, Anfosso F, Alessi MC, Nalbone G, Lafont H,
Juhan-Vague I. Stimulating effect of oxidized low density lipoproteins on
plasminogen activator inhibitor-1 synthesis by endothelial cells. Arterioscler Thromb. 1991;11:1821–1829.
12. Tremoli E, Camera M, Maderna P, Sironi L, Prati L, Colli S, Piovella F,
Bernini F, Corsini A, Mussoni L. Increased synthesis of plasminogen
activator inhibitor-1 by cultured human endothelial cells exposed to
native and modified LDLs: an LDL receptor–independent phenomenon.
Arterioscler Thromb. 1993;13:338 –346.
13. Kugiyama K, Sakamot T, Misumi I, Sugiyama S, Ohgushi M, Ogawa H,
Horiguchi M, Yasue H. Transferable lipids in oxidized low-density
lipoprotein stimulate plasminogen activator inhibitor-1 and inhibit
tissue-type plasminogen activator release from endothelial cells. Circ Res.
1993;73:335–343.
14. Karikó K, Rosenbaum H, Kuo A, Zurier RB, Barnathan ES. Stimulatory
effect of unsaturated fatty acids on the level of plasminogen activator
inhibitor-1 mRNA in cultured human endothelial cells. FEBS Lett. 1995;
361:118 –122.
15. Boberg M, Pollare T, Siegbahn A, Vessby B. Supplementation with n-3
fatty acids reduces triglycerides but increases PAI-1 in non–insulindependent diabetes mellitus. Eur J Clin Invest. 1992;22:645– 650.
16. Moller JM, Svaneborg N, Lervang H-H, Varming K, Madsen P, Dyerberg
J, Schmidt EB. The acute effect of a single very high dose of n-3 fatty
acids on coagulation and fibrinolysis. Thromb Res. 1992;67:569 –577.
17. Eritsland J, Arnesen H, Seljeflot I, Kierulf P. Long-term effects of n-3
polyunsaturated fatty acids on haemostatic variables and bleeding
episodes in patients with coronary artery disease. Blood Coagul Fibrinolysis. 1994;6:17–22.
18. Oosthuizen W, Vorster HH, Jerling JC, Barnard HC, Smuts CM, Silvis N,
Kruger A, Venter CS. Both fish oil and olive oil lowered plasma
fibrinogen in women with high baseline fibrinogen levels. Thromb
Haemost. 1994;72:557–562.
19. Eriksson P, Nilsson L, Karpe F, Hamsten A. Very-low-density lipoprotein
response element in the promoter region of the human plasminogen
activator inhibitor-1 gene implicated in the impaired fibrinolysis of hypertriglyceridemia. Arterioscler Thromb Vasc Biol. 1998;18:20 –26.
20. Panahloo A, Mohamed-Ali V, Lane A, Green F, Humphries SE, Yudkin
JS. Determinants of plasminogen activator inhibitor-1 activity in treated
NIDDM and its relation to a polymorphism in the plasminogen activator
inhibitor-1 gene. Diabetes. 1995;44:37– 42.
21. Mansfield MW, Stickland MH, Grant PJ. Environmental and genetic
factors in relation to elevated circulating levels of plasminogen activator
inhibitor-1 in Caucasian patients with non–insulin-dependent diabetes
mellitus. Thromb Haemost. 1995;74:842– 847.
22. Ossei-Gerning N, Mansfield MW, Stickland MH, Wilson IJ, Grant PJ.
Plasminogen activator inhibitor-1 promoter 4G/5G genotype and plasma
levels in relation to a history of myocardial infarction in patients characterized by coronary angiography. Arterioscler Thromb Vasc Biol. 1997;
17:33–37.
November 1998
1685
23. Edgell C-JS, McDonald CC, Graham JB. Permanent cell line expressing
human factor VIII-related antigen established by hybridization. Proc Natl
Acad Sci U S A. 1983;80:3734 –3737.
24. Karpe F, Steiner G, Olivecrona T, Carlson LA, Hamsten A. Metabolism
of triglyceride-rich lipoproteins during alimentary lipemia. J Clin Invest.
1993;91:748 –758.
25. Spector AA, Hoak JC. An improved method for the addition of long chain
free fatty acid to protein solutions. Anal Biochem. 1969;32:297–302.
26. Lund E, Diczfalusy U, Björkhem I. On the mechanism of oxidation of
cholesterol at C-7 in a lipoxygenase system. J Biol Chem. 1992;267:
12462–12467.
27. Bradford MM. A rapid and sensitive method for the quantification of
microgram quantities of protein utilizing the principle of protein-dye
binding. Anal Biochem. 1976;72:248 –254.
28. Sambrook J, Fritsch EF, Maniatis T. Molecular Cloning: A Laboratory
Manual. 2nd ed. Cold Spring Harbor, NY: Cold Spring Harbor Laboratory; 1989.
29. Alksnis M, Barkhem T, Strömstedt P-E, Ahola H, Kutoh E, Gustafsson
J-Å, Poellinger L, Nilsson S. High level expression of functional full
length and truncated glucocorticoid receptor in Chinese hamster ovary
cells. J Biol Chem. 1991;266:10078 –10085.
30. Kalb VF, Bernlohr RW. A new spectrophotometric assay for protein in
cell extracts. Anal Biochem. 1977;82:362–371.
31. Crofts JW, Ogburn PL, Johnson SB, Holman RT. Polyunsaturated fatty
acids of serum lipids in myocardial infarction. Lipids. 1988;23:539 –545.
32. Green S, Wahli W. Peroxisome proliferator-activated receptors: finding
the orphan a home. Mol Cell Endocrinol. 1994;100:149 –153.
33. Issemann I, Prince RA, Tugwood JD, Green S. The peroxisome
proliferator-activated receptor:retinoid X receptor heterodimer is activated by fatty acids and fibrate hypolipidaemic drugs. J Mol Endocrinol.
1993;11:37– 47.
34. Göttlicher M, Widmark E, Li Q, Gustafsson J-Å. Fatty acids activate a
chimera of the clofibric acid-activated receptor and the glucocorticoid
receptor. Proc Natl Acad Sci U S A. 1992;89:4653– 4657.
35. Keller H, Dreyer C, Medin J, Mahfoudi A, Ozato K, Wahli W. Fatty acids
and retinoids control lipid metabolism through activation of peroxisome
proliferator-activated receptor-retinoid x receptor heterodimers. Proc Natl
Acad Sci U S A. 1993;90:2160 –2164.
36. Krey G, Braissant O, L’Horset F, Kalkhoven E, Perroud M, Parker MG,
Wahli W. Fatty acids, eicosanoids, and hypolipidemic agents identified as
ligands of peroxisome proliferator-activated receptors by coactivatordependent receptor ligand assay. Mol Endocrinol. 1997;11:779 –791.
37. Elkeles RS, Chakrabarti R, Vickers M, Stirling Y, Meade TW. Effect of
treatment of hyperlipidaemia on haemostatic variables. BMJ. 1980;281:
973–974.
38. Simpson HC, Mann JI, Meade TW, Chakrabarti R, Stirling Y, Woolf L.
Hypertriglyceridaemia and hypercoagulability. Lancet. 1983;1:786 –790.
39. Andersen P, Nilsen DWT, Lyberg-Beckmann S, Holme I, Hjermann I.
Increased fibrinolytic potential after diet intervention in healthy coronary
high-risk individuals. Acta Med Scand. 1988;223:499 –506.
40. Sundell IB, Dahlgren S, Rånby M, Lundin E, Stenling R, Nilsson TK.
Reduction of elevated plasminogen activator inhibitor levels during
modest weight loss. Fibrinolysis. 1989;3:51–53.
41. Folsom AR, Qamhieh HT, Wing RR, Jeffery RW, Stinson VL, Kuller
LH, Wu KK. Impact of weight loss on plasminogen activator inhibitor
(PAI-1), factor VII, and other hemostatic factors in moderately overweight adults. Arterioscler Thromb. 1993;13:162–169.
42. Schuit AJ, Schouten EG, Kluft C, de Maat M, Menheere PP, Kok FJ.
Effect of strenuous exercise on fibrinogen and fibrinolysis in healthy
elderly men and women. Thromb Haemost. 1997;78:845– 851.
43. Lopez-Segura F, Velasco F, Lopez-Miranda J, Castro P, Lopez-Pedrera
R, Blanco A, Jimenez-Pereperez J, Torres A, Trujillo J, Ordovas JM,
Pérez-Jiménez F. Monounsaturated fatty acid-enriched diet decreases
plasma plasminogen activator inhibitor type 1. Arterioscler Thromb Vasc
Biol. 1996;16:82– 88.
44. Toft I, Bonaa KH, Ingebretsen OC, Nordoy A, Jenssen T. Fibrinolytic
function after dietary supplementation with omega-3 polyunsaturated
fatty acids. Arterioscler Thromb Vasc Biol. 1997;17:814 – 819.
Downloaded from http://atvb.ahajournals.org/ by guest on June 18, 2017
Unsaturated Fatty Acids Increase Plasminogen Activator Inhibitor-1 Expression in
Endothelial Cells
Lennart Nilsson, Cristina Banfi, Ulf Diczfalusy, Elena Tremoli, Anders Hamsten and Per
Eriksson
Arterioscler Thromb Vasc Biol. 1998;18:1679-1685
doi: 10.1161/01.ATV.18.11.1679
Arteriosclerosis, Thrombosis, and Vascular Biology is published by the American Heart Association, 7272
Greenville Avenue, Dallas, TX 75231
Copyright © 1998 American Heart Association, Inc. All rights reserved.
Print ISSN: 1079-5642. Online ISSN: 1524-4636
The online version of this article, along with updated information and services, is located on the
World Wide Web at:
http://atvb.ahajournals.org/content/18/11/1679
Permissions: Requests for permissions to reproduce figures, tables, or portions of articles originally published
in Arteriosclerosis, Thrombosis, and Vascular Biology can be obtained via RightsLink, a service of the
Copyright Clearance Center, not the Editorial Office. Once the online version of the published article for which
permission is being requested is located, click Request Permissions in the middle column of the Web page
under Services. Further information about this process is available in thePermissions and Rights Question and
Answer document.
Reprints: Information about reprints can be found online at:
http://www.lww.com/reprints
Subscriptions: Information about subscribing to Arteriosclerosis, Thrombosis, and Vascular Biology is online
at:
http://atvb.ahajournals.org//subscriptions/